Is knowledge just justified true belief?
For two thousand years, philosophers said knowledge is a true belief you have good reason for. In 1963, a three-page paper showed that you can have all three and still not know.
- Reading time
- 5 min
- Difficulty
- Intermediate
- Branch
- Knowledge
- First posed
- Edmund Gettier, 'Is Justified True Belief Knowledge?', Analysis, 1963; anticipated in Plato, Theaetetus
The setup
What is it to know something? A popular answer, often traced to Plato's Theaetetus, is that knowledge is justified true belief. To know that it is raining, you must believe it is raining; it must actually be raining (you cannot know what is false); and you must have good reason to believe it, rather than a lucky guess.
In 1963 a little-known philosopher named Edmund Gettier, reportedly pressed by his colleagues to publish something, wrote a paper less than three pages long. It contained two counterexamples, and it changed epistemology overnight.
Here is a simpler case in the same spirit. You look at a clock in the hall. It says three o'clock, so you believe it is three o'clock. You have good reason: the clock has always been reliable. And it is three o'clock. But, unknown to you, the clock stopped exactly twelve hours ago. You have a justified, true belief. Do you know what time it is?
Almost everyone says no. You were lucky. Your belief is true, and justified, but it is true by accident, not because of your justification. If that is right, justified true belief is not enough for knowledge.
Why it matters
The Gettier problem launched decades of attempts to find the missing ingredient in knowledge. It is a model of how a single clear counterexample can overturn a long-standing theory.
It also matters wherever reliability and luck come apart: in science, where a correct result can come from a flawed experiment; in law, where a conviction can be right for the wrong reasons; and in everyday life, where we often believe true things on misleading grounds.
Key terms
- Justified true beliefThe traditional analysis of knowledge: you know something when you believe it, it is true, and you are justified in believing it. Gettier cases challenge it.
- Gettier problemThe challenge, raised by Edmund Gettier in 1963, of explaining why a justified true belief that is true only by luck does not count as knowledge.
- Epistemic luckLuck that makes a belief true in a way that seems incompatible with knowing it. Gettier cases are the classic examples.
- ReliabilismThe view that a belief is justified, or counts as knowledge, if it is produced by a reliable process, one that usually produces true beliefs.
- JustificationHaving good reasons or evidence for a belief, so that it is reasonable to hold.
The positions
Add a fourth condition
Knowledge is justified true belief plus something else that rules out luck.
Reliabilism
Knowledge is true belief produced by a reliable process. Justification, in the traditional sense, is not required.
Knowledge is unanalysable
Knowledge cannot be broken down into simpler parts. It is basic, and the search for a definition is misguided.
Virtue epistemology
You know when your true belief is a success that comes from your intellectual competence, not from luck.
Arguments and objections
Add a fourth condition
- Best argument
- Many proposals have been made. The 'no false lemmas' view requires that your justification not depend on any false belief: in the clock case, you relied on the false belief that the clock was working. 'Defeasibility' views require that there be no truth which, if you learned it, would undermine your justification. Each handles some Gettier cases.
- Strongest objection
- Each proposal has met new counterexamples. In the fake barn case, you look at a real barn in a county full of convincing barn facades; your belief that it is a barn depends on no false belief, yet many say you do not know. Fifty years of patches have not produced a consensus.
Reliabilism
- Best argument
- Alvin Goldman proposed that what matters is whether a belief was produced by a reliable cognitive process, like normal perception in good conditions. A stopped clock is not a reliable process for telling the time, even if it happens to be right. This externalist view does not require you to be aware of why your belief is reliable.
- Strongest objection
- Processes can be described more or less broadly ('looking at a clock' versus 'looking at this stopped clock'), and reliabilism struggles to say which description counts. This is called the generality problem.
Knowledge is unanalysable
- Best argument
- Timothy Williamson argued in Knowledge and Its Limits (2000) that knowledge is the most general factive mental state, and is prior to belief and justification, not built from them. The failure of every attempt to analyse it after Gettier is evidence that the project was wrong from the start.
- Strongest objection
- If knowledge cannot be analysed, we lose a way of explaining why it matters and why it differs from lucky true belief. Many epistemologists still believe some analysis is possible.
Virtue epistemology
- Best argument
- Ernest Sosa and John Greco compare knowledge to an archer's shot that hits the target because of the archer's skill, not because of a lucky gust of wind. In Gettier cases, your belief is true, and you were competent, but the truth is not due to your competence.
- Strongest objection
- It is not always clear what makes a success 'due to' competence. In the fake barn case, your competent perception did produce your true belief, yet many still deny that you know.
Who argued what
- Add a fourth condition Michael Clark, Keith Lehrer, Peter Klein
- Reliabilism Alvin Goldman, Fred Dretske
- Knowledge is unanalysable Timothy Williamson
- Virtue epistemology Ernest Sosa, John Greco, Linda Zagzebski
Where you stand
Can knowledge be defined?
In surveys
Choose one to see how people answer.
Variations
Gettier's original case
Smith and Jones have applied for a job. Smith has strong evidence that Jones will get it, and that Jones has ten coins in his pocket. He infers: 'the man who will get the job has ten coins in his pocket.' In fact, Smith himself gets the job, and, unknown to him, Smith also has ten coins in his pocket. His belief is justified and true, but he does not know it.
The sheep in the field
Roderick Chisholm described seeing what looks like a sheep in a field, and believing there is a sheep in the field. It is actually a dog disguised as a sheep. But behind a rock, out of view, there is a real sheep. Your belief is true and justified, but it is about the wrong animal.
The fake barns
Driving through a county, you look at a barn and think 'that's a barn'. It is. But unknown to you, the county is full of fake barn facades, and you happened to look at the only real one. Do you know it is a barn? Intuitions here are more divided, and the case was used to test many theories.
Go deeper
Older cases
Gettier was not the first to notice the problem. The eighth-century Indian philosopher Dharmottara described a traveller who sees what looks like water on the road, actually a mirage, and correctly believes there is water there, because there happens to be water under a rock. Bertrand Russell gave the stopped clock case in 1948. But it was Gettier's paper that made the problem impossible to ignore.
Related ideas
Further reading and sources
- EncyclopediaStanford Encyclopedia of Philosophy: The Analysis of Knowledge
- WikipediaGettier problem
- SurveyPhilPapers Survey, 2020 · question: Analysis of knowledge; about 1,785 professional philosophers; accept or lean toward
- PaperIs Justified True Belief Knowledge?, Edmund Gettier
- BookKnowledge and Its Limits, Timothy Williamson
- BookEpistemology: A Very Short Introduction, Jennifer Nagel
Open questions
- Can you know something by luck?
- If you can never rule out luck entirely, do you know anything?
- Does it matter whether you know something, if what you believe is true?